© Borgis - New Medicine 4/2006, s. 97-101
*Irena Maniecka-Bryła1, Ireneusz Kuropka2
Increase in life expectancy as an expression of favourable changes in mortality*
1Department of Social and Preventive Medicine, Medical University in Łódź
Head of Department: Prof. Wojciech Drygas, M.D., Ph.D.
2Department of Economic Forecasts and Analyses, Academy of Economics in Wrocław
Head of Department: Prof. Paweł Dittmann, Ph.D
Summary
Summary
Introduction. This paper aims to evaluate the changes in overall mortality and life expectancy of Łódź inhabitants, with an emphasis on the 65-74 age group, which occurred during 11 years of the socio-economic transformation and demographic transition in Poland.
Material and methods. Our study is based on secondary sources. We make use of complete mortality databases of the Łódź population, the WHO database, data tables of the Central Statistical Office stored in an archive, and materials of the Regional Public Health Centre in Łódź. Our analysis concerns two years: 1991 and 2002. We applied the following statistical methods: indices of proportion, intensity and dynamics, direct standardisation, u-test for two frequencies. In order to calculate life expectancies in the investigated age groups in Łódź in 1991 and 2002, we used the method proposed by Greville.
Results. Favourable changes in the life expectancy profile for every year in the 65-74 age group among both genders in Łódź were observed. At the same time, the differential of male and female life expectancies diminished in 2002 in comparison with 1991. It turned out that the length of life of a male Łódź inhabitant aged 65 was shorter by 0.77 year in 1991 than that of a male Pole in the same age category, whereas it was 1.13 year for a female inhabitant compared to a female Pole. The elimination of deaths due to CVD would have resulted in a gain of 9.92 years in e(65) in 1991 among males and 10.33 years among females, whereas in 2002 it would have been 4.48 and 4.20 years respectively.
Conclusions. Positive changes in mortality were confirmed. Mortality is considered to be an important element of the health situation of the "young old” (i.e. aged 65-74) population in Łódź in the period under study, which brought about an increase in survival probability. The magnitude of the gain in life expectancy, assuming a total elimination of CVD mortality, justifies our opinion that these diseases constitute the most important health problem.
Introduction
This paper aims to evaluate the changes in overall mortality and life expectancy of Łódź inhabitants, with an emphasis on the 65-74 age group, which occurred during 11 years of the socio-economic transformation and demographic transition in Poland.
We chose the year 1991 as the reference point of our considerations due to the worst epidemiological situation, in particular as far as the most important health problem, i.e. cardiovascular diseases (CVD), is concerned. The health situation of the population considered to constitute the young old population in 1991 was compared to that in 2002, when clearly positive changes of health phenomena examined by us had continued for several years. In the period under study, apart from the socio-economic and epidemiological transitions, various demographic changes occurred, the result of which was, inter alia, the progressive process of our city population´s ageing [1, 2]. It is worth noting that in this period, the number of Łódź inhabitants decreased by 52,239 people, i.e. by 6.26%, and reached 782,540 on 30 June 2002. The number of inhabitants aged 65 and over increased by 13,693, i.e. by 11.73%, and reached 130,409 in 2002. People aged 65-74 constituted the biggest group of the population aged 65 and over – 64.30% in 1991 and 59.15% in 2002, i.e. respectively 75,401 and 77,131 people. This means that in the period under study, every tenth Łódź inhabitant belonged to this age group [3]. Because of the unfavourable demographic and health situation of Łódź inhabitants compared to the entire country, this paper provides numerous references to Poland as a whole, particularly while evaluating the most important mortality elements.
Material and methods
Our study is based on secondary sources. We make use of complete mortality databases of the Łódź population, the WHO database, data tables of the Central Statistical Office stored in an archive, and materials of the Regional Public Health Centre in Łódź. Our analysis concerns two years: 1991 and 2002. We applied the following statistical methods: indices of proportion, intensity and dynamics, direct standardisation, u-test for two frequencies [4]. In order to calculate life expectancies in the investigated age groups in Łódź in 1991 and 2002, we used the method proposed by Greville. Since this method has rarely been used in medical papers, it seems desirable to describe it in general terms.
Life tables are a collection of biometric functions showing the order of ageing of a hypothetical generation. The basic parameter of life tables – the probability of death – is calculated on the basis of real partial mortality indices occurring in a given period. The other variables are derived from the calculated death probabilities for a given age [5, 6, 7].
Therefore, life expectancy at x measures the life potential of a person of a given age. Life expectancy at birth describes in a synthetic way mortality conditions concerning a population in a given period. Despite being a hypothetical value, it is the most common measure used to evaluate this kind of situation. This parameter reflects the impact of various factors shaping the mortality of a population and it depends on the conditions of life. It is affected by many social, economic and cultural factors [8, 9, 10, 11, 12]. That is why E. Rosset [13, 14] considered life expectancy to be "a barometer of social progress”.
The method of calculating life table parameters based on the assumption of an even distribution of deaths is called traditional and it is seldom used in practice. The reason is above all the changeable and uneven distribution of infant deaths. To calculate the death probabilities, a modification of the formula proposed by Greville, allowing a reduction of this error [15], was used.
Result
Due to the limits of this paper, we will present only the extension of life expectancy for the 65-74 age group in the period 1991-2002 as well as an assessment of gains in this important health situation indicator on the assumption of a total elimination of CVD mortality.
Table 1 shows the dynamics of overall mortality in Łódź compared to Poland in the period under study. Data included in this table indicate that overall mortality indices both in Łódź and in Poland declined, but they were always significantly higher in Łódź than in Poland. The dynamics of the decline of the overall mortality rate in Łódź was similar to that in Poland, although it continues to be lower than in Łódź (by 3.6% in 2002).
Table 1. The dynamics of overall mortality in Łódź and in Poland in the period 1991-2002.
| Year | Łódź | Poland |
| rate per 1000 population | 1991=100 | rate per 1000 population | 1991=100 |
| 1991 | 14.6 | 100.00 | 10.5 | 100.00 |
| 1992 | 14.3 | 97.95 | 10.3 | 98.10 |
| 1993 | 14.2 | 97.26 | 10.2 | 97.14 |
| 1994 | 14.2 | 97.26 | 10.0 | 95.24 |
| 1995 | 14.1 | 97.26 | 10.0 | 95.24 |
| 1996 | 13.9 | 96.57 | 10.0 | 95.24 |
| 1997 | 13.9 | 95.21 | 9.8 | 93.33 |
| 1998 | 13.9 | 95.21 | 9.7 | 92.38 |
| 1999 | 14.3 | 97.95 | 9.9 | 94.29 |
| 2000 | 13.8 | 94.52 | 9.6 | 91.43 |
| 2001 | 13.8 | 94.52 | 9.5 | 90.48 |
| 2002 | 13.0 | 89.04 | 9.4 | 89.52 |
Source: own research on the basis of data provided by the Regional Statistical Office in Łódź and http://www.3.who.int/whosis/mort/table1_process.cfn.
Due to the more advanced ageing process in Łódź than in the entire Polish population, so as to assess the overall mortality rate differentials which result from the health situation and not from the demographic situation, direct standardisation was carried out. This enabled us to make the compared indices independent of population age structure differences. With the aid of standardisation, one may establish how many deaths would have occurred in the analysed population (Łódź in this case) if the age structure had been identical to that of a standard population. We took the structure of the Polish population as a standard. Our calculated standardised rates are much lower than the real ones among males, females and in the entire population. The differential between Łódź and Poland in 1991 fell to 1.2% in the total population and to 0.8% in 2002. However, overall mortality rates in Łódź continue to be statistically significantly higher than in Poland at the significance level of p<0.001, which was confirmed by a u-test (Figs. 1 and 2). This is evidence for our thesis that the health situation of Łódź inhabitants is worse than that of Poles, because mortality rates are its most negative measures. Comparing overall mortality rates in Łódź and in Poland in the years under study, we found that they were statistically significantly (p <0.001) higher in Łódź both among males and females. In the period under study, there was a favourable reduction of these rates both in the Łódź population and in that of Poland. The standardised overall mortality rate in Łódź fell by 12.8% (14.7% among males and 11.5% among females). In Poland, this decline was smaller and it amounted to: 10.5%, 11.1% and 10.5% respectively. It is worth noting that the dynamics of decline of overall mortality rates in the population aged 65-74 was higher than the dynamics of decline of these rates for the entire population. This applies to both males and females. In Łódź, this rate fell by 21.2% (18.1% among males and 22.6% among females), whereas in Poland it fell by 17.3%, 15.1% and 22.0% respectively.
In 1991, deaths in the 65-74 age group constituted 24.59% of all deaths in Łódź (compared to 21.35% in Poland). In 2002, the share in Łódź was slightly lower (at 23.70%), while it was 24.60% in Poland. The number of deaths in this group in Łódź fell from 2994 in 1991 to 2413 in 2002, i.e. by 581. Together with the population growth in this age group, it caused positive changes in mortality rates (Table 2).
Table 2. Overall mortality by age and sex in Łódź in 1991 and 2002.
| Age group | All deaths (rates per 1000 population) |
| 1991 | 2002 |
| Total | Males | Females | Total | Males | Females |
| N | rate | n | rate | n | rate | N | rate | n | rate | n | rate |
| 0-4 | 117 | 2.92 | 65 | 3.18 | 52 | 2.66 | 62 | 2.29 | 36 | 2.59 | 26 | 1.98 |
| 5-14 | 30 | 0.27 | 18 | 0.32 | 12 | 0.22 | 16 | 0.23 | 12 | 0.33 | 4 | 0.12 |
| 15-24 | 72 | 0.73 | 51 | 1.03 | 21 | 0.43 | 53 | 0.47 | 34 | 0.60 | 19 | 0.34 |
| 25-34 | 205 | 1.84 | 152 | 2.85 | 53 | 0.91 | 121 | 1.13 | 95 | 1.78 | 26 | 0.49 |
| 35-44 | 776 | 4.86 | 552 | 7.28 | 224 | 2.67 | 361 | 3.51 | 259 | 5.28 | 102 | 1.89 |
| 45-54 | 978 | 10.27 | 696 | 15.72 | 282 | 5.53 | 1162 | 7.95 | 807 | 12.05 | 355 | 4.48 |
| 55-64 | 2197 | 21.63 | 1457 | 34.20 | 740 | 12.55 | 1329 | 15.63 | 866 | 23.42 | 463 | 9.63 |
| 65-74 | 2994 | 39.71 | 1647 | 56.42 | 1347 | 29.15 | 2413 | 31.28 | 1313 | 46.17 | 1100 | 22.59 |
| 75-84 | 3252 | 97.44 | 1283 | 124.01 | 1969 | 85.50 | 3064 | 68.67 | 1234 | 89.84 | 1830 | 59.26 |
| 85-94 | 1455 | 191.10 | 387 | 229.67 | 1068 | 180.13 | 1453 | 177.63 | 392 | 202.17 | 1061 | 170.00 |
| 95+ | 99 | 303.68 | 13 | 203.13 | 86 | 328.24 | 148 | 307.69 | 20 | 235.29 | 128 | 323.23 |
| Total | 12175 | 14.58 | 6321 | 16.43 | 5854 | 13.01 | 10182 | 13.01 | 5068 | 14.16 | 5114 | 12.04 |
Source: own calculations.
Table 3 presents the structure of deaths by disease classes in the 65-74 age group in Łódź in the years under study. Attention should be drawn to the decline in the share of deaths due to CVD among both genders as well as the rise in the share of deaths caused by malignant neoplasms.
Table 3. Structure of deaths in the 65-74 age group in Łódź in 1991 and 2002.
| Disease class | Share in all deaths (in %) |
| 1991 | 2002 |
| Males | Females | Males | Females |
| Cardiovascular diseases | 53.19 | 55.98 | 38.61 | 41.09 |
| Malignant neoplasms | 23.80 | 23.09 | 31.68 | 31.09 |
| Digestive system diseases | 4.98 | 4.31 | 4.42 | 5.73 |
| Respiratory system diseases | 4.80 | 3.04 | 5.64 | 4.00 |
| External death causes | 2.55 | 2.82 | 1.90 | 1.00 |
| Infectious and parasitic diseases | 0.55 | 0.37 | 0.46 | 0.18 |
| Other causes | 10.13 | 10.39 | 17.29 | 16.91 |
| Total | 100.00 | 100.00 | 100.00 | 100.00 |
Source: own calculations.
Now we would like to make reference to the synthetic measure of the health situation, i.e. life expectancy. This is the best known parameter of life tables. It expresses the average number of years to be lived in given mortality conditions by a member of the investigated population who has completed x years of life. This measure synthesises all the factors which influence the mortality level of a given population. That is why it is often used in comparisons of the health situation of various populations. It is also used to assess a change in a period of time. Life expectancy at birth (e(0)) is widely considered to be one of the most accurate mortality measures, and its level in a given population is often used to evaluate social progress [16].
In the examined period, i.e. from 1991 to 2002, very favourable changes in the value of this synthetic health situation measure occurred in our country. The difference in life expectancy at birth between females and males was 9.16 years in 1991 and 8.36 years in 2002. How was e(0) for Łódź inhabitants evolving against the background of Poland? In 1955-56 e(0) was slightly higher than in Poland (+0.2 years). Among females e(0) was 68.3 years, and the difference was bigger (it amounted to +1.3 years). In the nineteen sixties, e(0) values continued to be more favourable in Łódź than in Poland for both males and females, as e(0) was 67.4 years (+1.5 year difference) and 74.1 years (+2.3 years) respectively in 1965-66. The situation changed in the seventies, when a new-born male Pole had a chance to survive 1 year longer than a male Łódź inhabitant (e(0)=65.7 years). For a female Pole, it was 0.4 year more than in Łódź (e(0)=73.2 years).
At the beginning of the 21st century this unfavourable situation (for Łódź) remains, as a male new-born in 2002 had a chance of surviving 68.0 years, which was over 2 years less than the average Pole (e(0)=70.3 years). A similar differential in life expectancy at birth for Łódź inhabitants compared to Poles remains, but is gradually diminishing. In 2004, e(0) in Łódź was 68.7 years among males and 77.8 among females, whereas e(0) in Poland was 70.7 and 79.2 years respectively. Table 4 shows the positive changes in life expectancy in the 65-74 age group among both genders in Łódź. The difference in life expectancy between females and males in Łódź in 2002 decreased compared to 1991.
Table 4. Life expectancy (in years) for people aged 65-75 in Łódź in 1991 and 2002.
| Year | 1991 | 2002 |
| Males | Females | Differential F-M | Males | Females | Differential F-M |
| 65 | 11.43 | 14.77 | 3.34 | 13.19 | 16.31 | 3.12 |
| 70 | 9.15 | 11.32 | 2.17 | 10.40 | 12.52 | 2.12 |
| 75 | 6.69 | 8.10 | 1.41 | 7.93 | 9.06 | 1.13 |
Source: own calculations.
The comparison of e(65) in Łódź and Poland in the selected years justifies the following statements. The length of life of a male Łódź inhabitant aged 65 in 1991 was lower than that of a Pole by 0.77 years. Regarding females, it was 1.13 years. In 2002, these differentials remained and amounted to 0.81 and 1.50 respectively. The dynamics of the rise of e(65) in selected years was similar in Łódź and in Poland. The increase of e(65) was higher among males (1.76 years) than females (1.54 years).
There arises a question: how significant would the increase in life expectancy in the 65-75 age group have been if deaths due to CVD, which constitute the most important mortality cause, had been eliminated? Table 5 presents our calculations of this gain in e(x) in the years under study. In this hypothetical situation, male e(65) for a Łódź inhabitant in 1991 would have been significantly longer, as it would have gained 9.92 years to reach 21.35 years. For females of the same age, the gain would have been bigger: 10.23 years, which means that their survival chances would have gone up to 25 years. Considerable increases in life expectancy would have occurred among both males and females aged 70 and 75 years as well. Referring to 2002, in spite of the enormous improvement of the situation in the field of CVD mortality reduction, which brought about an extension of e(x), the gains due to total elimination of deaths caused by CVD in the 65-75 age group would still have been very significant. For 65-year-olds, it would have brought an extension of life expectancy by 4.48 years among males and 4.20 years among females; for 70-year-olds, 3.86 and 3.62 years.
Table 5. Gain in life expectancy of those aged 65-75 by sex, assuming complete elimination of CVD mortality in 1991 and 2002.
| Age | 1991 | 2002 |
| Males | Females | Males | Females |
| 65 | 9.92 | 10.23 | 4.48 | 4.20 |
| 70 | 9.77 | 10.05 | 4.27 | 3.96 |
| 75 | 9.64 | 9.46 | 3.86 | 3.62 |
Source: own calculations.
The situation described above is an entirely theoretical concept, as there are no methods permitting complete eradication of CVD mortality, but it certainly confirms their health significance. Furthermore, it makes us aware of the huge potential of life expectancy extension connected with CVD mortality reduction thanks to primary and secondary prevention measures [17, 18, 19, 20].
Conclusions
1. Positive changes in mortality were confirmed. Mortality is considered to be an important element of the health situation of the "young old” (i.e. aged 65-74) population in Łódź in the period of socio-economic transformation under study, which resulted in an increase in survival probability.
2. The magnitude of the gain in life expectancy, assuming total eradication of CVD mortality, justifies our thesis that these diseases constitute the most important health problem among the elderly.
*Our research (including this publication) was financed by the Medical University in Łódź in the framework of grant no. 502-16-516.
Piśmiennictwo
1. Leowski J. Transformacja demograficzna i epidemiologiczna - nowe wyzwania zdrowia publicznego. [W:] Leowski J. red. Polityka zdrowotna a zdrowie publiczne. CeDeWu, Warszawa 2004, s. 41-50. 2.Kurek S. Przestrzenne zróżnicowanie starzenia się ludności w Polsce w układzie miast i gmin w okresie 1989-1998. [W:] Kowaleski J.T., Szukalski P., red. Proces starzenia się ludności - potrzeby i wyzwania. Wydawnictwo Uniwersytetu Łódzkiego, Łódź 2002, s. 96-105. 3.Maniecka-Bryła I., Bryła I., Szymańska I. Tendencje zmian umieralności w populacji Łodzi ze szczególnym uwzględnieniem grupy wiekowej 65-74 lata. [W:] Kowaleski J.T., Szukalski P., red. Starość i starzenie się jako doświadczenie jednostek i zbiorowości ludzkich. Uniwersytet Łódzki, Łódź 2006, s. 226-231. 4.Maniecka-Bryła I., Martini-Fiwek J., red. Epidemiologia z elementami biostatystyki, Wydawnictwo Uniwersytetu Medycznego w Łódzi, Łódź 2005. 5.Bolesławski L. Budowa tablic trwania życia. Teoria i praktyka. Studia i Prace Statystyczne, nr 47. GUS, Warszawa 1973. 6.Bolesławski L. Prognoza trwania życia osób starszych. [W:] Kuciarska-Ciesielska M., Marciniak G., red. Seniorzy w polskim społeczeństwie. Zakład Wydawnictw Statystycznych. Warszawa 1999, s. 26-28. 7. Holzer J. Demografia. Państwowe Wydawnictwo Ekonomiczne. Warszawa 1980. 8.Kuropka I. Zgony. [W:] Cieślak M., red. Demografia. Metody analizy i prognozowania. Wydawnictwo Naukowe PWN. Warszawa 1992. 9.Kuropka I. Natężenie zgonów według wieku jako czynnik przyrostu trwania życia ludności Dolnego Śląska w latach 1980-1998. Studia Demograficzne, 1999; 2. 10.Kuropka I. Umieralność osób starszych a trwanie życia mieszkańców Dolnego Śląska w latach 1980-1998. [W:] Kowaleski J., Szukalski P., red. Sytuacja zdrowotna osób w starszym wieku w Polsce. Aspekt medyczny i społeczno-demograficzny II. Oficyna Wydawnicza Instytutu Medycyny Pracy, Łódź 2000, s. 83-93. 11.Kuropka I. Zmiany natężenia zgonów osób starszych a długość życia mieszkańców Polski w latach dziewięćdziesiątych. [W:] Kowaleski J., Szukalski P., red.. Proces starzenia się ludności - potrzeby i wyzwania. Wydawnictwo Uniwersytetu Łódzkiego, Łódź 2002, s. 203-213. 12.Kuropka I. Potencjał życiowy mieszkańców Dolnego Śląska. Diagnoza i perspektywy. Akademia Ekonomiczna we Wrocławiu, Wrocław 2002. 13.Rosset E. Tablice wymieralności jako barometr postępu społecznego. Ekonomista, 1959; 4-5. 14.Rosset E. Ludzie starzy. Studium demograficzne. PWE, Warszawa 1967, 467 s. 15.Greville T. Short methods of constructing abridged life tables. Records of the American Institute of Actuaries, 1943; 32. 16.Okólski M. Demografia. Podstawowe pojęcia, procesy i teorie w encyklopedycznym zarysie, Wyd. Naukowe SCHOLAR, Warszawa 205, 312 s. 17.Bryła M, Maniecka-Bryła I, Stelmach W, Stelmach I. Ocena sytuacji epidemiologicznej w zakresie chorób układu krążenia w populacji województwa łódzkiego. Wiadomości Lekarskie, 2002; 15(supl. 1, cz. 2): 639-645. 18.Maniecka-Bryła I, Bryła M. Choroby układu krążenia - najważniejszy problem zdrowotny ludzi starych na przykładzie województwa łódzkiego. [W:] Kowaleski J., Szukalski P., red. Nasze starzejące się społeczeństwo. Nadzieje i zagrożenia. Wydawnictwo Uniwersytetu Łódzkiego. Łódź 2004: 134-140. 19.Maniecka-Bryła I, Bryła M, Stelmach W. Sytuacja epidemiologiczna w zakresie chorób układu krążenia w województwie łódzkim jako uzasadnienie dla realizacji programów profilaktycznych. Annales Academiae Medicae Stetinensis, 2005; 51(supl. 1): 97-103. 20.Maniecka-Bryła I, Bryła M, Drygas W. Sytuacja epidemiologiczna chorób układu krążenia w województwie łódzkim na tle sytuacji epidemiologicznej w Polsce na początku XXI wieku. Przegląd Epidemiologiczny, 2005; 59: 923-932.
